Damian Sojka, Yuyang Liu, Dipam Goswami, Sebastian Cygert, Bartłomiej Twardowski, & Joost van de Weijer. (2023). Technical Report for ICCV 2023 Visual Continual Learning Challenge: Continuous Test-time Adaptation for Semantic Segmentation.
Abstract: The goal of the challenge is to develop a test-time adaptation (TTA) method, which could adapt the model to gradually changing domains in video sequences for semantic segmentation task. It is based on a synthetic driving video dataset – SHIFT. The source model is trained on images taken during daytime in clear weather. Domain changes at test-time are mainly caused by varying weather conditions and times of day. The TTA methods are evaluated in each image sequence (video) separately, meaning the model is reset to the source model state before the next sequence. Images come one by one and a prediction has to be made at the arrival of each frame. Each sequence is composed of 401 images and starts with the source domain, then gradually drifts to a different one (changing weather or time of day) until the middle of the sequence. In the second half of the sequence, the domain gradually shifts back to the source one. Ground truth data is available only for the validation split of the SHIFT dataset, in which there are only six sequences that start and end with the source domain. We conduct an analysis specifically on those sequences. Ground truth data for test split, on which the developed TTA methods are evaluated for leader board ranking, are not publicly available.
The proposed solution secured a 3rd place in a challenge and received an innovation award. Contrary to the solutions that scored better, we did not use any external pretrained models or specialized data augmentations, to keep the solutions as general as possible. We have focused on analyzing the distributional shift and developing a method that could adapt to changing data dynamics and generalize across different scenarios.
|
Dani Rowe. (2008). Towards Robust Multiple-Target Tracking in Unconstrained Human-Populated Environments.
|
Dani Rowe, Ignasi Rius, Jordi Gonzalez, & Juan J. Villanueva. (2005). Robust Particle Filtering for Object Tracking.
|
Dani Rowe, Ignasi Rius, Jordi Gonzalez, & Juan J. Villanueva. (2005). Improving Tracking by Handling Occlusions.
|
Daniel Marczak, Grzegorz Rypesc, Sebastian Cygert, Tomasz Trzcinski, & Bartłomiej Twardowski. (2023). Generalized Continual Category Discovery.
Abstract: Most of Continual Learning (CL) methods push the limit of supervised learning settings, where an agent is expected to learn new labeled tasks and not forget previous knowledge. However, these settings are not well aligned with real-life scenarios, where a learning agent has access to a vast amount of unlabeled data encompassing both novel (entirely unlabeled) classes and examples from known classes. Drawing inspiration from Generalized Category Discovery (GCD), we introduce a novel framework that relaxes this assumption. Precisely, in any task, we allow for the existence of novel and known classes, and one must use continual version of unsupervised learning methods to discover them. We call this setting Generalized Continual Category Discovery (GCCD). It unifies CL and GCD, bridging the gap between synthetic benchmarks and real-life scenarios. With a series of experiments, we present that existing methods fail to accumulate knowledge from subsequent tasks in which unlabeled samples of novel classes are present. In light of these limitations, we propose a method that incorporates both supervised and unsupervised signals and mitigates the forgetting through the use of centroid adaptation. Our method surpasses strong CL methods adopted for GCD techniques and presents a superior representation learning performance.
|
Daniel Marczak, Sebastian Cygert, Tomasz Trzcinski, & Bartlomiej Twardowski. (2023). Revisiting Supervision for Continual Representation Learning.
Abstract: In the field of continual learning, models are designed to learn tasks one after the other. While most research has centered on supervised continual learning, recent studies have highlighted the strengths of self-supervised continual representation learning. The improved transferability of representations built with self-supervised methods is often associated with the role played by the multi-layer perceptron projector. In this work, we depart from this observation and reexamine the role of supervision in continual representation learning. We reckon that additional information, such as human annotations, should not deteriorate the quality of representations. Our findings show that supervised models when enhanced with a multi-layer perceptron head, can outperform self-supervised models in continual representation learning.
|
Daniel Ponsa, A.F. Sole, Antonio Lopez, Cristina Cañero, Petia Radeva, & Jordi Vitria. (2000). Regularized EM..
|
Daniel Ponsa, A.F. Sole, Antonio Lopez, Cristina Cañero, Petia Radeva, & Jordi Vitria. (1999). Regularized EM.
|
Daniel Ponsa, & Antonio Lopez. (2009). Seguimiento Visual de Contornos Computerizado.
|
Daniel Ponsa, Antonio Lopez, Felipe Lumbreras, Joan Serrat, & T. Graf. (2005). 3D Vehicle Sensor based on Monocular Vision.
|
Daniel Ponsa, Antonio Lopez, Joan Serrat, Felipe Lumbreras, & T. Graf. (2005). Multiple Vehicle 3D Tracking Using an Unscented Kalman Filter.
Keywords: vehicle detection
|
Daniel Ponsa, & Jordi Vitria. (1999). Mobile monitoring system using an agent-oriented approach.
|
Daniel Ponsa, & Xavier Roca. (2002). Unsupervised Parameterisation of Gaussian Mixture Models.
|
Daniel Ponsa, & Xavier Roca. (2002). A Novel Approach to Generate Multiple Shape Models..
|
David Geronimo, Angel Sappa, Antonio Lopez, & Daniel Ponsa. (2006). Pedestrian Detection Using AdaBoost Learning of Features and Vehicle Pitch Estimation.
Abstract: In this paper we propose a combination of different Haar filter sets and Edge Orientation Histograms (EOH) in order to learn a model for pedestrian detection. As we will show, with the addition of EOH we obtain better ROCs than using Haar filters alone. Hence, a model consisting of discriminant features, selected by AdaBoost, is applied at pedestrian-sized image windows in order to perform
the classification. Additionally, taking into account the final application, a driver assistance system with realtime requirements, we propose a novel stereo-based camera pitch estimation to reduce the number of explored windows.
With this approach, the system can work in urban roads, as will be illustrated by current results.
Keywords: ADAS, pedestrian detection, adaboost learning, pitch estimation, haar wavelets, edge orientation histograms.
|